Skip to content

RunningTests

Jump to bottom
Zearin edited this page Jan 12, 2012 · 3 revisions

Running Tests

Testify has a powerful and flexible test runner. But out of the box, it's pretty simple to use.

    rhettg@devvm1:~/src/Testify $ bin/testify test
    ...............
    PASSED.  15 tests / 12 cases: 15 passed (0 unexpected), 0 failed (0 expected).  (Total test time 0.00s)

Running a specific test

If you add a test runner to your test case files themselves:

    if __name__ == "__main__":
      run()

You can then just execute the file:

    rhettg@devvm1:~/src/Testify $ python test/test_case_test.py
    ...............
    PASSED.  15 tests / 11 cases: 15 passed (0 unexpected), 0 failed (0 expected).  (Total test time 0.00s

However, using the testify runner is more powerful, and easier to remember:

    rhettg@devvm1:~/src/Testify $ bin/testify test.test_case_test TestMethodsGetRun
    ..
    PASSED.  2 tests / 1 case: 2 passed (0 unexpected), 0 failed (0 expected).  (Total test time 0.00s

Running Suites

Assuming you have collected tests together using Suites, it's easy to mix and match:

    rhettg@devvm1:~/src/Testify $ bin/testify test -i basic -x disabled

This will run all the tests in the basic suite, excluding any tests that have been marked disabled (by putting them in the disabled suite)

Exclusion suites can be very helpful for when you have tests that shouldn't be run in certain environments. For example we have a set of tests that require some number of external services to be setup, so if you include those tests in the requires-services suite, you have control over whether they are run or not.

Additional Reporting

Testify can be easily extended (ExtendingTestify) for whatever your reporting needs are, but out of the box we already have a few options.

Running with --verbose option gives you a rundown of all the tests and times being run:

    rhettg@devvm1:~/src/Testify $ bin/testify test --verbose
    test.test_case_test ClassSetupFixturesGetRun.test_test_var ... ok in 0.00s
    test.test_case_test DeprecatedClassSetupFixturesGetRun.test_test_var ... ok in 0.00s
    test.test_case_test DeprecatedClassTeardownFixturesGetRun.test_placeholder ... ok in 0.00s
    test.test_case_test DeprecatedSetupFixturesGetRun.test_test_var ... ok in 0.00s
    .....

    PASSED.  15 tests / 12 cases: 15 passed (0 unexpected), 0 failed (0 expected).  (Total test time 0.00s)

In addition, by using the --summary option all test failures are collected in a easy to use report at the end of run

    rhettg@devvm1:~/src/Testify $ bin/testify test --verbose --summary
    test.test_case_test ClassSetupFixturesGetRun.test_test_var ... ok in 0.00s
    test.test_case_test DeprecatedClassSetupFixturesGetRun.test_test_var ... ok in 0.00s
    test.test_case_test DeprecatedClassTeardownFixturesGetRun.test_placeholder ... ok in 0.00s
    test.test_case_test DeprecatedSetupFixturesGetRun.test_test_var ... ok in 0.00s
    .....
    =======================================================================
    FAILURES
    The following tests are expected to pass.

    ========================================================================
    test.test_case_test TestMethodsGetRun.test_method_1
    ------------------------------------------------------------
    Traceback (most recent call last):
      File "./test/test_case_test.py", line 5, in test_method_1
        assert_equal(1 + 1, 3)
      File "/nail/home/rhettg/src/Testify/testify/assertions.py", line 32, in assert_equal
        assert lval == rval, "assertion failed: %r == %r" % (lval, rval)
    <type 'exceptions.AssertionError'>: assertion failed: 2 == 3
    ========================================================================

    ========================================================================
    test.test_case_test TestMethodsGetRun.assert_test_methods_were_run
    ------------------------------------------------------------
    Traceback (most recent call last):
      File "./test/test_case_test.py", line 13, in assert_test_methods_were_run
        assert self.test_1_run
    <type 'exceptions.AttributeError'>: 'TestMethodsGetRun' object has no attribute 'test_1_run'
    ========================================================================

    FAILED.  16 tests / 12 cases: 14 passed (0 unexpected), 2 failed (0 expected).  (Total test time 0.00s)

In addition to normal stdout test output, we've also got a component for storing results in JSON format for later machine processing:

    rhettg@devvm1:~/src/Testify $ bin/testify test --json-results=results.json --extra-json-info="{\"tag\": \"production\"}"
    rhettg@devvm1:~/src/Testify $ head -1 results.json
    {"name": "testify.test_case DeprecatedSetupFixturesGetRun.classSetUp", "success": true, "start_time": 1281141702.0, "module": "testify.test_case", "tag": "production", "run_time": 1.7e-05, "type": "fixture", "end_time": 1281141702.0}

Bucketing and Parallelization

Using the options --bucket-count and --bucket allows you to split up your test set into separate sets that can be run individually.

The buckets are determined deterministically, so a given test case will always end up in the same bucket given the same number of buckets.

For large projects, this can be very helpful to spread out tests across multiple machines or multiple cores.

Clone this wiki locally